perf: add threading capability to perfbdiff

Since we are releasing the GIL during diffing, it is interesting to see how a
thread pool would perform on diffing. We add a new `--threads` argument to
commands. Synchronizing the thread pool is a bit complex because we want to be
able to reuse it from one run to another.

On my computer (i7 with 4 cores + hyperthreading), I get the following data for
about 12000 revisions:
threads wall        comb    wall gain   comb overhead
none    31.596715   31.59    0.00%       0.00%
1       31.621228   31.62   -0.08%       0.09%
2       16.406202   32.8    48.08%       3.83%
3       11.598334   34.76   63.29%      10.03%
4        9.205421   36.77   70.87%      16.40%
5        8.517604   42.51   73.04%      34.57%
6        7.94645    47.58   74.85%      50.62%
7        7.434972   51.92   76.47%      64.36%
8        7.070638   55.34   77.62%      75.18%

Compared to the feature disabled (threads=0), the overhead is negligible with
the threading code (threads=1), and the gain is already 48% with two threads.

contrib/perf.py

@@ -25,7 +25,9 b' import os'
 import random
 import struct
 import sys
+import threading
 import time
+import util.queue
 from mercurial import (
     changegroup,
     cmdutil,
@@ -933,11 +935,25 b' def perffncacheencode(ui, repo, **opts):'
     timer(d)
     fm.end()
 
+def _bdiffworker(q, ready, done):
+    while not done.is_set():
+        pair = q.get()
+        while pair is not None:
+            mdiff.textdiff(*pair)
+            q.task_done()
+            pair = q.get()
+        q.task_done() # for the None one
+        with ready:
+            ready.wait()
+
 @command('perfbdiff', revlogopts + formatteropts + [
     ('', 'count', 1, 'number of revisions to test (when using --startrev)'),
-    ('', 'alldata', False, 'test bdiffs for all associated revisions')],
+    ('', 'alldata', False, 'test bdiffs for all associated revisions'),
+    ('', 'threads', 0, 'number of thread to use (disable with 0)'),
+    ],
+
     '-c|-m|FILE REV')
-def perfbdiff(ui, repo, file_, rev=None, count=None, **opts):
+def perfbdiff(ui, repo, file_, rev=None, count=None, threads=0, **opts):
     """benchmark a bdiff between revisions
 
     By default, benchmark a bdiff between its delta parent and itself.
@@ -983,14 +999,39 b' def perfbdiff(ui, repo, file_, rev=None,'
             dp = r.deltaparent(rev)
             textpairs.append((r.revision(dp), r.revision(rev)))
 
+    withthreads = threads > 0
+    if not withthreads:
     def d():
         for pair in textpairs:
             mdiff.textdiff(*pair)
-
+    else:
+        q = util.queue()
+        for i in xrange(threads):
+            q.put(None)
+        ready = threading.Condition()
+        done = threading.Event()
+        for i in xrange(threads):
+            threading.Thread(target=_bdiffworker, args=(q, ready, done)).start()
+        q.join()
+        def d():
+            for pair in textpairs:
+                q.put(pair)
+            for i in xrange(threads):
+                q.put(None)
+            with ready:
+                ready.notify_all()
+            q.join()
     timer, fm = gettimer(ui, opts)
     timer(d)
     fm.end()
 
+    if withthreads:
+        done.set()
+        for i in xrange(threads):
+            q.put(None)
+        with ready:
+            ready.notify_all()
+
 @command('perfdiffwd', formatteropts)
 def perfdiffwd(ui, repo, **opts):
     """Profile diff of working directory changes"""

tests/test-contrib-perf.t

@@ -175,7 +175,7 b' Check perf.py for historical portability'
   $ (testrepohg files -r 1.2 glob:mercurial/*.c glob:mercurial/*.py;
   >  testrepohg files -r tip glob:mercurial/*.c glob:mercurial/*.py) |
   > "$TESTDIR"/check-perf-code.py contrib/perf.py
-  contrib/perf.py:498:
+  contrib/perf.py:\d+: (re)
    >     from mercurial import (
    import newer module separately in try clause for early Mercurial
   [1]